There is need to develop an efficient and accurate method to predict in-season N requirements for winter wheat in the southeast. The Illinois Soil Nitrogen Test (ISNT) has been used with some success in the mid-west to determine N fertilizer recommendations for row crops such as corn. The objective of this research is to develop an in-season fertilizer N recommendation system using INST for winter wheat in the humid southeast and to determine the effect of time and depth of sampling on the INST values. Field experiments were conducted under conventional and no-till conditions during the 2005-2006 growing season at 23 sites across North Carolina. A randomized complete block design with 5 replications was used with N treatment as the main plot factor. There were 7 fertilizer N treatments ranging from to 0 to 168 kg ha^-1. Soil samples were collected at planting and GS25 to a depth of 30 cm and analyzed for ASN using the procedure described by Khan et. al (2001). The fertilizer N treatments were applied at GS-30. Grain yield responses to N applications at GS-30 were modeled as linear-plateau function to determine the optimum N rate. Linear regression analysis was used to determine the relationships between optimum N rate determined at GS-30 and the INST values measured at planting and GS-25. Initial analysis indicates that the ISNT values are relatively stable from planting to GS 25. Preliminary results on the effects of sampling time and depth on the INST values will be discussed in further detail.